Interesting GBP/USD MA Crossover system

Thanks to everyone for your input and thoughts over these last 24 hours. There have been some questions asked.

MarkGavalda, how often to optimize? how large of a sample to optimize over? My beginning benchmark has been once a week over the last 6 months. As I make the EA better, we should be to optimize it over longer and longer periods. I believe though that the market wave's height and length will be substantially different from the waves 4-5 from now. So until I have a expert that can basically optimize itself continuously, I wouldn't optimize it over 8 months. But I don't want curve-fitting. When I optimize, I make the majority of the passes gave still gave me significant returns, and that there many many passes with similar parameters that got similar returns.

When the next version comes out, I'll make sure to add all the additions to it. I spent today moving the declaring and defining above the start(), and then getting rid of all the ranging code. Before I had crudely just doubled all the trend code for the ranging side of the EA, doubling the size of the EA, and making the EA define 80 parameters with every tick. It was extremely crude. So I got ride of all the range code, made all the code standard, (so now the EA has 40 parameters) and I move the defining above the start() function so it's only being called(and define) once for every bar. The end result is optimizations are much much quicker and the EA is much easier to work with since it's only half as long. Thanks a million to raytracy, I'd have no idea any of that was even possible without him.

Tonight I'm going to be loosening up on the entry conditions in some areas. With the range and trend trades I'm looking for at least 6-10 trends a month and we're not there yet. This trade that was opened this morning looks like its about to be closed with a loss. I don't know whether the market is going up or down tomorrow, but the EA should, so obviously its still not as smart as it should be.

Guys I've been watching the price for the last 6 hours and I've gotten some great ideas, I'm excited. To determine ranging, I'm adding the possibility if the average distance between the high and low of the last 8 candles is greater than the width between the 2 main EMAs.

Sounds great, but please include as many extern variables as possible, that way it will be easier to find the correct values with optimization.

but the more variables you have, the more prone you are to over-optimizing. I'd like to keep the number of variables to no more than 5, and have the system determine the rest.

hello aiccia,

FIrstly, thanks for posting this EA ! It looks good.

I downloaded the EA and ran it to see what it does. My understanding is that is a moving average EA (based on the crosses + the OSNA indicators). What I don't quite understand is that there are no SL or TP set in the orders ? SO how does this EA close the trade ?

Also with the optimisation, the EA will perform well because you have optimised it, but what about for example if you optimise it for the Jan and then using those settings see how it performs in Feb ?

Cheers, Anthony

You can have a look at that yourself you said you had downloaded and ran the EA already.
Optimization is for finding better values for an already profit-making EA, I didn't believe this and a week (or two) ago I started to find out if it's true for myself, and yes it is.
The big question is that which period should you use when optimizing and for how long will those values work?

aicccia!
You're right about that, by the way I started an MQL4 topic in the MetaTrader 4 sub-forum, it's about auto-optimization. Have a look and comment if you have time! (That goes to all of you, of course!)

Have a good weekend!

For info., my moneymaker just opened a buy on GBP/USD at 1.9374 (demo on alpari). We'll see the result...

I was up 56 pips and after lunch (CET) I'm down 55pips ... I knew I had to close my position!

I suppose I should give some sort of an update. I haven't been posting here because I've been working on this dumb EA which I'm getting sick and tired of. I've worked on it 6 hours and counting today, 8 hours yesterday, maybe 4 hours on saterday, and a similar rate for the last 2 weeks. The huge headache has been getting the EA to switch back and forth between timeframes, trying to figure out why your metatrader suddenly stops backtesting, spending a day or two going crazy only to realize that it was issues you'd only run into doing 30,000+ optimization passes an hour(so no one else had idea what your talking about), running out of memory, huge logs files, running out of hard drive space, figuring out how get around these problems and still optimize that much, try again and again and again and again to get the EA to switch timeframes using only one variable as your timeframe, then giving up and making two complete identical function defining all your parameters with the only difference being Period_H1 for the period for one function and Period_M15 for the period for the other function, finally getting this to all work and then realizing this switching back forth of the timeframes slows the time of one pass from 1 second to 10-15+ seconds, and optimization times go from 30 minutes to 16 hours, trying this once and realizing this is not going to work because it will take me a week to try out just 4 ideas if I have a problem(which I always do), and then FINALLY giving up on ONE EA and splitting everything into two EAs; MoneyMaker-Trend as one and MoneyMaker_Range as the other.

So you see, I'm getting sick and tired of anything to do with metatrader. BUT, I'm almost done. I can see what to do. Before, I was just guessing, trying it, guessing again and trying another thing because the coding was just beyond my skill level.

So I'm almost done. I know I said that a week ago but this time I mean it. thanks for waiting. it will be worth it.

yes. yesterday was a ranging market. it would not have happened with the versions coming.

with the new versions, it would have bought around 1.9338, closed at 1.9402, sold at 1.9389 and closed at 1.9300, for around 150 pips. Approximately, Approximately, but you get the general idea, it would have caught both moves beautifully. It's very hard to make something work that well, thats why its taking me longer than expected, forgive me.

But let me say this again, the new version,first complete version will actually be two EAs, on two charts.

Of course if there's someone really good at MQ4 out there who'd want to combine it all very well so it would still optimize fast, I'd be eternally grateful because then the optimizing and backtesting could be more efficient and accurate.

Or maybe not, maybe this is all a blessing in disguise. It will be easier to fit the ranging just right if it's a separate EA.

Many thanks for the great work, it's amazing. Take a good rest and clear your mind if you need! Everybody here is searching for the best EA for a long long time ago so if we have to wait one more week it's not that bad!
thx again

noo I don't want to rest, I want it to be done!

I'd say the new version would put in a buy for GBP/USD at 1.9296. we can see where it would go. This is just a guess though made on default values for my EMAs, without them being optimized.

Anyone listen to Enya?
Imagine her softly chanting....

"Go up, Go up, Go up!"

yeah I'm pretty sure I'm going crazy now, time for some sleep, time for some sleep...goodnight forex world

I was a professional programmer once (web development), I can understand what you're talking about, mate.
Your EA looks promising, we can easily tell that you have a very smart brain, so don't give up, this kind of work will have its rewards sooner or maybe later, but for sure.
I'm really looking forward to your next version, I have some Core2duo processor time if you need some for speeding up the optimizations.

Aicccia,

Nothing to forgive. You have done an oustanding job and we all applaud you!

Many thanks for taking so much time and effort to formulate what has been a Herculean effort. The actual possibility of designing an EA to work in both Ranging and Trend markets is nothing short of miraculous!

Thank you again for such a great job...

Theo

New Update.

Sadly, but predictably the 15 minute timeframe has about as much in common with the hourly charts as a house cat does with a cheetah. I was uneducated enough to think I could apply my hourly, trending code to the much more unpredictable 15 minutes and with just a few tweaks and optimizing, get it to work up to my standards. After about a day I realized I was all the way back at square one with this new system. I had just spent months working on the GBP/USD system for the hourly charts and I felt I was going to have to start all over and make a whole new system, very demoralizing.

Now the whole reason I need another system is so I could be in the market when it was ranging for two reasons, in this order: 1)be in the market in case a small range trend develops into a large hourly trend, 2)to make money when on the hourly charts the price is just moving sideways. So though I already do have a fully working, functional hourly system, its not good enough for me because I know I could make 2x as much money and get at least a third more trades with a 2nd 15 minute mini-system working in concert with the hourly one.

Fast forward a week of frustrating dead-ends and I've had several good ideas, but either the drawdown was too high(around 20% for fixed lot MM) or the optimization produced illogical results. And that's a big thing for me, I don't care how profitable a system is, if it's not logical, I'm not trading it. So I wasn't happy, not happy at all, and I was in the very frustrating part of system development when I'm looking for something but I have no idea what it looks like, so I'm reduced to guessing and trying random ideas. It's like old school non-genetic optimization.

So anyway, I was playing around with my osama indicator and becoming more familiar with the effects of changing the parameters (it has 5). By the way, it has a few very interesting, almost magical characteristics. For example if you make the Panda_Fast parameter equal to 1 and the Panda_Slow and Panda_Signal the same number, then enter and exit when it crosses 0 then in a ranging market and it always gets in and out at around the same price. So I was hunting around around for more magical properties and when I realized I could get it to cross 0 the same time of my jumps using the ma lines.

This was huge. Remember half of my reason for needing a ranging 15 minute system was because on the hourly markets the two MA lines got too close together and I start having bad jumps constantly. With these settings, the osama cleanly crosses once even in a severe ranging market. And, even more importantly this meant I could get rid of every bit of code that was related to MA lines, making the whole EA much more simple, and much more logical. In fact I got rid of over 75% of the EA. The spike detector is gone, the jump trades, waterfall trades, and spring trades are all gone, and all their supporting code. The EA was around 500 lines, now it's less than 150 and it's much more precise and accurate.

So I've definitely made another huge leap this week in improving the MoneyMaker.

Hey Aicccia,
I havent yet tried your system, but its clear to me you're putting a great effort into this project. Therefore, I must say I really admire what you are doing, and I'm definitely gonna be following your progress. Keep it coming!
Matt

Aiccia -

Looks like you're tuning that lemon into lemonade. Congratulations!

Hello Aicccia... When I came across your nickname... and noticed your mind goes by "Nothing is Impossible"... I wanted to learn more about who this person is... red the post... and find your thread...

You are an amazing individual... I am convinced you will never give up... you'll get what you want... and what you want will always evolve... so you will be always vibrant and young person till the end of your days on this earthly trip... Congratulations!

I am still a pilgrim without a house [in this forex adventure]... so am glad to find your open house... and your round table guests... I am not a programmer... my demo MT4 [mig]... so I could only test your EA there... I know, its not much, but that's all what I can do for now... am looking forward to install and test your new versions...

Thank YOU... and ALL... for sharing... thank you for being there...